End of training

.gitattributes

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
 trainer_log.jsonl filter=lfs diff=lfs merge=lfs -text
+trainer_state.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # train_qnli_101112_1760638087
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the qnli dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0421
+- Loss: 0.0383
 - Num Input Tokens Seen: 207147488
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.038302481174468994,
+    "eval_runtime": 218.51,
+    "eval_samples_per_second": 47.938,
+    "eval_steps_per_second": 11.986,
+    "num_input_tokens_seen": 207147488,
+    "total_flos": 9.327757760604144e+18,
+    "train_loss": 0.030981469616896774,
+    "train_runtime": 93516.2877,
+    "train_samples_per_second": 20.161,
+    "train_steps_per_second": 5.04
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "eval_loss": 0.038302481174468994,
+    "eval_runtime": 218.51,
+    "eval_samples_per_second": 47.938,
+    "eval_steps_per_second": 11.986,
+    "num_input_tokens_seen": 207147488
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "num_input_tokens_seen": 207147488,
+    "total_flos": 9.327757760604144e+18,
+    "train_loss": 0.030981469616896774,
+    "train_runtime": 93516.2877,
+    "train_samples_per_second": 20.161,
+    "train_steps_per_second": 5.04
+}

trainer_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbd69b6d59328d66ba5ecf28fac5008636c4eb7c8c71ebb29be688bd9aff019f
+size 19947707